SETI@Homepage

I was trying to figure out how you could potentially make a SETI@Home client implemented in Javascript. With some quick Googling I found an interesting proof of concept paper describing: “Unwitting Distributed Genetic Programming via Asynchronous JavaScript and XML“.

We present a proof-of-concept system for
distributed computation of genetic programming via asyn-
chronous javascript and XML (AJAX) techniques which re-
quires no explicit user interaction and no installation of
client side software. Clients automatically and possibly even
unknowingly participate in a distributed genetic program-
ming system simply by visiting a webpage, thereby allowing
for the solution of genetic programming problems without
running a single local fitness evaluation. The system can be
easily introduced into existing webpages to exploit unused
client-side computation for the solution of genetic program-
ming and other problems.

Which I thought lent some interesting quasi legitimacy to my idea, only to almost immediately have my all these hopes dashed by the SETI@Home problem. Essentially the question is: How Can You Verify the results of a Distributed Calculation.

Potentially well meaning, developers have sought to “improve” the underlying SETI@Home algorithm to make it faster and to gain massive nerd bragging rights.
Unfortunately, the side effect of this also risks introducing subtle and overall extremely disruptive statistical anomalies into the overall project.

So, to combat well meaning, but misguided ideas like mine, the SETI project has implemented some classic methods of verifying distributed computations.

What is most interesting with all the the different techniques is that you get a real sense of a vibrant computational ecosystem of checks and cross checks happening within the staid confines of what is essentially a giant math problem.